Cycling Assay for Nicotinamide. Adenine. Dinucleotide. CARL BERNOFSKY AND. MARCELLA SWAN. Laboratory of Molecular Biology,. Mayo. Clinic and Mayo ...
ANALYTICAL
BIOCHEMISTRY
An
Improved
53, 452-458 (1973)
Cycling Adenine
CARL BERNOFSKY Laboratory
of Molecular
Biology, Rochester,
Assay for Dinucleotide AND
Nicotinamide
MARCELLA
Mayo
Minnesota
SWAN
Clinic and Mayo Foundation, 55901
ReceivedAugust 30, 1972; accepted January 3, 1973 A new cycling assay for NAD that uses thiazolyl blue as a terminal electron acceptor has been found to offer significant advantages over the more established procedure that employs 2,6dichlorophenolindophenol. With thiazolyl blue, the cycling assay is linear with NAD at picomole levels, and with time for at least 120 min. In contrast, with 2,6-dichlorophenolindophenol as terminal acceptor, the cycling assay deviates considerably from linearity at picomole levels of NAD, and the reaction rates become linear for shorter periods of time as the level of NAD increases. Data are given which provide a basis for choosing optimal assay conditions using the new thiazolyl blue cycling technique.
Recent studies of pyridine nucleotides in muscle extracts have involved the isolation of the bound forms of NAD+ and NADH by means of ultrafiltration (1). Because the NADH content of rabbit muscle is small [4050 nmoles/g fresh tissue (1) ] and only limited amounts of extract can be ultrafiltered, a highly sensitive assay technique is required. A study of the suitability of the cycling procedure of Slater et al. (2) revealed that it was insufficiently sensitive as well as nonlinear in the required range. Nisselbaum and Green (3) have recently proposed the use of thiazolyl blue as a terminal electron acceptor in the cycling assay for pyridine nucleotides. We have found that a modified version of their method possessesboth the required sensitivity and the linearity for the analysis of ultramicro levels of bound NAD+ and NADH. The present report describes this modified thiazolyl blue technique. We also discuss certain important features of the assay that have not been previously noted. METHODS AND MATERIALS Chemicals. Thiazolyl blue and Bicinel were from Calbiochem, PES was from Nutritional Biochemicals Corp., and Tris, 2,6-DCPIP, iodoacetate, ; ’ The following abbreviations are used: Bicine, N.N-bis(2-hydroxyethyl)glycine 2,6-DCPIP, 26dichlorophenolindophenol; PES, 5-ethylphenazinium ethyl sulfate ; PMS, 5-methylphenazinium methyl sulfate. 452 Copyright @ 1973 by Academic Press, Inc. All rights of reproduction in any form reserved.
IMPROVED
CYCLING
ASSAY
FOR
453
NAD
EDTA, NAD+, PMS, and crystalline yeast alcohol dehydrogenase (Type 34O-26)? were obtained from Sigma Chemical Co. All other chemicals were of reagent grade quality, and solutions were prepared with glass-distilled water that was previously deionized. Aqueous solutions (unbuffered) of PES and PMS could be stored in absolute darkness for up to 1 year at room temperature wit,ho~t loss of catalytic potency. Assay Co~~i~~o~. All manipulations involving assay media were performed in subdued light with a minimum of delay. The compositions of the assay media varied in different experiments and are listed in the legends to the figures, Samples containing NAD+ in 13 X 65 mm tubes were mixed with assay medium and preincubated in the dark at 30°C to establish thermal equilibrium. The volume of the reaction mixture was 1.2 ml, and, unless otherwise noted, assays were initiated by addition of alcohol dehydrogenase. Alcohol dehydrogenase was reconstituted, before use, with 40 mM Tris, 40 rnM KH2P0,, 20 mM Na,P,O, (final pH ~8.1) and kept on ice. Changes in absorption were followed in a l.O-cm light path, 0.25-ml capacity flow cell (Thomas, Type 8495-1110; A. H. Thomas Co., Phila., PA) inst.alled in a Gilford Model 2000 spectrophotometer which was equipped with a temperature-regulated sample compartment kept at 30°C. RESULTS
Comparison of Cycling Procedures for NAD. Initially, an attempt was made to improve the sensitivity of the conventional cycling procedure of Slater et al. (2) by increasing the concentrations of alcohol dehydrogenase and PES above those previously used ]compare legend to Fig. 1 with previous conditions (1) 1. Although the rates of P,6-DCPIP reduction are greater with the higher levels of alcohol ~lehydrogenase and PES, they are nonlinear with respect to NAD+ at low concentrations of NAD+ (Fig. 1). Moreover, the rates of indophenol reduction are not linear with time at NAD’ concentrations of 0.1 nmole/assay or higher. In the latter case, the rate of indophenol reduction (not shown) typically increases during the initial 4 min of the reaction, remains constant until about 55% of the 2,6-DCPIP is consumed, then begins to decline, presumably because the electron acceptor has become rate-limiting. Although it would seem that the terminal electron acceptor could be prevented from becoming rate-limiting by simply increasing its initial concentration, the amount of oxidized 2,6-DCPIP that can be present is limited by the
‘Type 340-26 alcohol been replaced by Type personal communication)
dehydrogenase A-1762, which .
is no longer available is depleted of endogenous
from Sigma. NAD (D.
It has Broida,
454
BERNOFSKY
0
0.04 NAD+
AND
0.00
SWAN
0.12
0.16
0.20
(nmolehsay)
FIG. 1. Comparison of cycling procedures for NAD. Each assay contained, in a volume of 1.2 ml, 72 Fmoles of Na Bicine (pH 7.8), 0.6 mmole of ethanol, 2 pmoles of PES, NAD’ as shown, and either 0.12 pmole of 2,6-DCPIP or 0.5 pmole of thiaaolyl blue. Assays were initiated with 10 ~1 (0.2 mg) of alcohol dehydrogenase, following a 5-min preincubation in the dark at 30°C. Changes in absorbance (decrease at 610 nm, increase at 570 nm) were recorded for 10 min. The means of triplicate determinations are shown. Data for indophenol reduction are based on the rapid, steady-state changes of absorbance.
photometric sensitivity of the spectrophotometer. In the present case, the initial concentration of 2,6-DCPIP is 0.12 pmole/assay (A,,, = 2). On the other hand, the oxidized form of thiazolyl blue is relatively colorless, and it is the appearance of optical density that is measured during the assay. This situation permits the use of greater concentrations of terminal acceptor, and, in the present case, 0.5 ymole/assay is used. As shown in Fig. 2, the absorption maximum of the reduced thiazolyl blue is 570 nm, a finding that was verified with samples of thiazolyl blue obtained from J. T. Baker Chemical Co. and Sigma Chemical Co. Nisselbaum and Green (3) previously reported the absorption maximum of reduced thiazolyl blue to be between 550 and 562 nm. Measurement of Picomole Levels. With a cycling time of 10 min and appropriate sensitivity of the spectrophotometer, as little as 10 pmoles of NAD may be satisfactorily determined by directly recording the rates of reduction of thiazolyl blue. However, smaller quantities of NAD require longer cycling times. The data in Fig. 3 show that the cycling assay with thiazolyl blue is linear with NAD’ from 1 to 10 pmoles/assay and with cycling times up to 120 min.
IMPRO~D
CYCLING ASSAY FOR NAD
500
600
wavelength
700
455
800
(t-m)
FIG. 2. Absorption spectrum of reduced thiazolyl blue. Each reaction mixture contained, in a volume of 1.0 ml, 120 pmoles of Na Bicine (pH 7.8), 0.6 mmole of ethanol, 2 pmoles of PES, 0.5 &mole of thiazolyl blue, and 0.25 mg of alcohol dehydrogenate. The assay was initiated, after a 5-min preincubation at 30°C by the addition of 0.2 ml of 1 BIG NAD’. The reaction proceeded in the dark for 15 min at 30°C and was terminated with 1.0 ml of 12 rnAd Na iodoacetate. For the control, 0.2 ml of water and 1.0 ml of 12 mM Na iodoacetate were added to the reaction mixture prior to incubation. Absorption spectra of both solutions were determined with a Beckman DK-2 spectrophotometer, using water as a reference.
Use of PES for PA&S. Substitution of PES for PMS does not significantly alter the rate of cycling. However, PES is chemically more stable than PMS, especially at higher pH’s, and this stability is important during long cycling times. McIlwain (4) showed that, under alkaline conditions, the methyl group of PMS is converted to formaldehyde in a reaction that is accompanied by an uptake of oxygen. Using an oxygen electrode to monitor the oxidative dealkylation of PM’S and PES (C. Bernofsky, unpublished data), it was found that, at pH 9.5, the rate of consumption of dissolved oxygen by 3.3 mM: PMS at 30°C is 73 FM 02/ min, which is ii-fold the rate observed when PES is used under identical conditions. These results confirm the relative stability of PBS over PMS.
466
BERNOFSKY
0
2
NAD+
4
6
(pmoles/assay)
8
IO
AND
SWAN
0
20
40
60
80
100
120
Minutes
FIG. 3. Assay of picomole levels of NAD. The reaction mixture contained, per ml, 120 pmoles of Na Bicine (pH 7.8), 0.6 mmole of ethanol, 2 pmoles of PES, 0.5 pmole of thiazolyl blue, and 0.2 mg of alcohol dehydrogenase. The mixture was equilibrated in the dark at 30°C without the enzyme. Alcohol dehydrogenase was then added, and the mixture was passed through a 0.45 brn Millipore filter. To initiate the assays, LO-ml aliquots were added to 0.2-ml samples containing from 0 to 10 pmoles of NAD’. The assays were incubated at 30°C for up to 120 min and terminated with 1.0 ml of 12 mM Na iodoacetate. Blanks containing no added NAD’ gave A,,,, values equivalent to 5 pmoles of NAD.
As a general rule, PMS should not be used at pH 8.5 or higher, nor PES at pH 9.5 or higher, for more than brief periods of time. Concentration Dependence of Alcohol Dehydrogenase and PES. The rate of reoxidation of NADH in the cycling assay is dependent upon the concentration of PES. However, at any fixed level of alcohol dehydrogenase, it is possible to approach a saturation point with respect to PES. In a similar fashion, the rate of reduction of NAD+ in the cycling assay is dependent on the concentration of alcohol dehydrogenase, and, at a fixed level of PES, one also approaches a saturation point with respect to alcohol dehydrogenase (Fig. 4). As a consequence of the mutual interdependence of PES and alcohol dehydrogenase in the cycling assay, it is virtually impossible to simultaneously saturate the system with both components. Thus, increasing the level of alcohol dehydrogenase increases the amount of PES required to saturate the assay, and increasing the level of PES increases the amount of alcohol dehydrogenase required to saturate the assay. The latter relationship is illustrated in Fig. 4 which shows that 0.25 mg/assay of alcohol dehydrogenase is near-saturating at a PES level of 0.5 pmole/ assay; however, the same amount of enzyme is far from saturating when the PES level is increased to 5 pmoles/assay. Optimization of Assay Conditions. In the procedure described by Nis-
IMPROVED CYCLING ASSAYFORNAD
451
24
h, ? 16 %
0
0.2 0.3 0.1 Alcohol dehydrugenase
0.4
0.5
(mg/assay)
FIG. 4. Saturation with alcohol dehydrogenase at different Ievels of PES. Each assay contained, in a volume of 1.2 ml, SO amoles of Na Bieine (pH 7.8), 0.6 mmolr of ethanol, 0.5 ;:mole of thiazolyl blue, 10 pmoles of NAD’, and from 0.2 to 5 pmoles of PES. Reactions were initiated, following preincubation in the dark at 3[Y”C, with 10 ~1 of alcohol dehydrogenase as shown. Changes in absorbance were recorded for 10 min.
selbaum and Green (3j, the final concentration of alcohol dehydrogenase in the cycling assay was 0.033 mg/ml. At this low level of alcohol dehydro~enase, the system was apparently saturated with PMS at 0.87 ,umole/ml. It should be noted that these conditions do not take full advantage of the sensitivity of which the assay system is capable. As indicated by the data in Fig. 4, sensitivity can be increased by elevating the concentrations of bot,h alcohol dehydrogenase and PES, and we have found that it is possible to achieve cycling rates that are IO- to 20-fold greater than those reported by Nisselbaum and Green (3). The practical upper limit of sensitivity appears to be imposed by the presence of enzyme-bound NADH. Bound NADH is found in varying degrees in commercial samples of crystalline yeast alcohol dehydrogenase, and this NADH is primarily responsible for the blank rate (C. Bernofsky and M. Swan, unpublished observations . The medium presently in routine use for the assay of 0.01-0.2 nmole of NAD+ or NADH (11 is the same as t.hat listed in the legend to Fig.
458
BE~NOFS~Y
AND
SWAN
DISCUSSION
The cycling assay of Slater et lal. (2)) which uses 2,6-DCPIP as a terminal electron, acceptor, presents a number of problems which limit its usefulness. One serious problem is the spontaneous reaction of 2,6DCPIP with thiols (mainly glutathione) that are present in cell extracts (5). In~rference from thiols can be eliminated by treating the extracts with appropriate amounts of N-ethylmaleimide (1). However, other troublesome characteristics of the assay are not so readily managed. These are the insensitivity and nonlinearity of the assay at low concentrations of NAD, and the fact that, because of spectrophotometric restrictions, only limited amounts of 2,6-DCPIP can be used in the cycling procedure. The latter limitation results in a shortening of the time period during which the cycling rate is linear at higher levels of NAD. It should be pointed out that the cycling procedure of Nisselbaum and Green (3) is also subject to interference by thiols (C. Bernofsky and M. Swan, unpublished observation). However, treatment of extracts with Nethylmaleimide is usually unnecessary, because the extracts can be sufficiently diluted to the point where thiol interference is negligible. The most useful features of the thiazolyl blue technique are that it is completely linear over a very wide range of NAD concentrations and time intervals and that the sensitivity of the assay can be readily manipulated by varying the concentrations of PES and alcohol dehydrogenase. These attributes, when considered together with the ease and simplicity of the method, make it seem likely that the thiazolyl blue procedure of Nisselbaum and Green (3), as modified by the present work, will probably supplant most other existing methods for the determination of pyridine nucleotides. ACKNOWLEDGMENTS This work was supported, in part, by Grant GB-26447 from the National Science Foundation and General Research Support Grant FR-5530 from the National Institutes of Health. REFERENCES 1. BRRNOFSKY, 2. SLATER, T.
C., AND PAN~OW, M. (1973) F., SAWYER, B., AND STR&JLI,
3. NXSSELZIAUM,
J. S., AND GREEN, S. (1969) Anal. Biochem.27,212. H. (1937) J. Chem. Sot. (London) 1704. C., AND ROYAL, K. M. (1970) Biochim. Biophgs.Acta
4. MCILWAIN, 6. BERNOFSICY,
Arch. B&o&m. &OphyS. U. (1964) Arch. Id.
lst;,
143.
Physiol. ~~o~h~rn.
215, 210.
